As the U.S. population ages, the health care system must change its emphasis from episodic acute illness treatment to a chronic illness management model to improve the quality, appropriateness, and effectiveness of health care services. Evidence-based guidelines and quality indicators have been developed to promote improvements in clinical practice for the vulnerable elderly; however primary care physicians (PCPs) rarely have the time, expertise, or incentive to provide the most effective evidence-based care. Referring these patients to the multiple specialists required to deal with the complex bio-psychosocial problems which often result in ineffective fragmented care and poor outcomes. Delivery of comprehensive care management using an interdisciplinary team (IT) that collaborates with the PCP has been shown to be an effective organizational model in chronic illness care. With this model, a care manager is usually assigned to provide patient follow up across all care settings and providers to ensure integration of medical and social issues. Use of this model to deliver healthcare services seems especially appropriate with frail elderly as a method to increase access to the expertise of the relatively few trained Geriatricians with the numerous PCPs who lack formal training in Geriatrics, yet care for increasing numbers of vulnerable older adults. Observational data from numerous care management programs indicate that this is a cost effective way to deliver optimal chronic illness care, however randomized trial evidence is lacking. This randomized trial will test the effectiveness of improved clinical practice through comprehensive care management in elderly patients with chronic illness and functional impairment discharged from an acute care hospital. For the intervention group, patient care will be coordinated by a nurse care manager who will perform a comprehensive in-home assessment and provide patient education and self-management support. The care manager will work with an IT to develop and implement a plan of care. Evidence based care plans will be implemented in collaboration with the patient, the PCP, the local Area Agency on Aging (AAoA), and other community social agencies. The care manager will provide frequent patient follow-up across all providers to ensure integration of medical and social issues. Control patients will be referred to the local AAoA with no IT follow up. Although control patients will receive, through the AAoA, referrals for care and psychosocial support, the absence of a care manager and IT will, we expect, result in functional decline, lower quality of life, and higher health care costs. The intervention (n=265) and control (n=265) groups will be compared at 1 year on a profile of health and well being using a multiple endpoint global hypothesis testing strategy. The global measure will be comprised of the following 5 domains: function, institutionalization, quality of life, quality of medical management and quality of self management. Priority populations identified by AHRQ who are targeted in this study include the elderly, patients with chronic illnesses, low income (dual eligible), and patients with disabilities. This study also includes minorities, women, and patients who live in the inner city. Cost effectiveness will be studied in a secondary analysis using incremental net benefit analysis. Economic analyses of benefits will inform policy makers about funding care management in AHRQ priority populations.